Buoyancy-driven Exchange Flow Research Articles

Colloidal exchange flow in ducts

Buoyancy-driven exchange flow, i.e. drainage of a heavy colloidal suspension by an immiscible light pure fluid in a vertical duct, is studied analytically. A lubrication model is developed for two-dimensional channel and axisymmetric pipe geometries. The outcome transport equations are solved numerically to capture the effect of important governing dimensionless parameters: Péclet number, precursor film thickness, Reynolds number, mixtures’ viscosity ratio, initial volume fraction of hard-sphere particles, particles concentration within the precursor film, and capillary number. Recent study of colloidal film flow over a flat surface reveals interface thinning over time due to diffusion impact on suspension viscosity. However, the current model suggests that there is no such thinning phenomenon observed for colloidal flows within confined geometry unless in the absence of surface tension. The extent of the mixed particle zone along the duct grows as the Péclet number is decreased. Precursor film thickness is found to decrease the capillary ridge height forming at heavy suspension front. Either an increase in the Reynolds number or a decrease in light-to-heavy-fluid viscosity ratio extends the interpenetration rate of mixtures. As the bulk volume fraction of particles is increased close to 50 %, interesting secondary fronts emerge within the flow that coincide with the range where the diffusion coefficient is minimized. The heavy suspension layer drains slightly faster when the gradient in particle concentration between the bulk and precursor film regions is maximum. Finally, the non-monotonic impact of surface tension on the extent of mixtures’ exchange zone is uncovered via simulations conducted at various capillary numbers.

Gravity currents over fixed beds of monodisperse spheres

Abstract

Influences of vent location on the removal of gaseous contaminants and indoor thermal environment

Heat sources and gaseous contaminants widely exist in modern buildings. Buoyancy-driven exchange flow through a large opening, such as a window, a doorway or an atrium, is ubiquitous in our daily life. Analogous experiments were carried out in a water tank to investigate the effect of vent location on the characteristics of the exchange flow. The model tank filled with coloured brine solution was immersed in a large reservoir filled with fresh water. The transient distribution of the reduced gravity is measured using a light attenuation technique. For the exchange flow through a vertical opening, the reduced gravity in the chamber is vertically stratified instead of uniformly distributed. Moreover, the density distribution was influenced by the location of the opening. The buoyancy-driven flow through a horizontal opening is highly oscillatory and the resulting density distribution is relatively uniform. Different models were proposed to estimate the ventilation flow rate through a vertical opening or a horizontal one. Hopefully, this study could provide useful advice with regards to the ventilation design and contaminant removal in buildings.

Bidensity particle-laden exchange flows in a vertical duct

Buoyancy-driven exchange flow in a vertical duct is studied theoretically for a light pure fluid and a heavy fluid. The latter is a suspension composed of a Newtonian fluid and two populations of negatively buoyant particles of the same size but different densities (bidensity). In a previous study (Mirzaeian & Alba, J. Fluid Mech., vol. 847, 2018, pp. 134–160), the authors developed a lubrication model for monodensity suspension of particles of uniform size. The main observation of the monodensity study was the discovery of particle-enriched zones near heavy and light fluid fronts due to the relative motion of particles and the fluid. Distinct from the previous work, here, mismatched densities instigate a relative motion of lighter and heavier particles in addition to the movement of fluids. Other than the previously observed enrichment case, the bidensity case gives rise to a novel flow regime where there is enrichment of heavy particles but depletion of light particles near the interface. The transition to this regime is governed by a balance between the densities of heavy and light particles as well as those of light and carrying fluids for a given choice of initial volume fractions of particles. Such density balance is characterized by two dimensionless parameters comprising light and heavy particle-to-carrying-fluid density ratios. The transition mechanism is studied through additional simulations, revealing that the former increases with initial volume fraction of particles of either type, while the latter contrarily decreases. The effect of other parameters on the flow are discussed within the context of the paper.

Buoyancy-driven unbalanced exchange flow through a horizontal opening

Buoyancy-driven exchange flows occur in a variety of natural and industrial situations, including nuclear and hydraulic engineering, oceanography and building ventilation. Balanced exchange flows, whereby there is simultaneously an equal volume flux transferred vertically upwards and downwards through a horizontal opening, have previously been described theoretically. However, until now there has been no theoretical description of unbalanced exchange flows, whereby the volume flux in one direction through an opening exceeds that in the other. The model developed herein examines the growth of perturbations on the density interface at an opening made in a horizontal plane that connects buoyant fluid below with denser fluid above. By considering the interface as it is advected away from the plane of the opening by a bulk flow imposed in the vertical, we quantify the exchange for the unbalanced case. The model successfully predicts the Froude number criterion, which corresponds directly to the minimum dimensionless flow rate of the imposed flow, for the onset of unbalanced exchange across circular openings found experimentally. Additionally, comparisons made between the exchanges predicted and measured show excellent agreement across the entire range of possible flows, from unidirectional flow, through unbalanced exchange to balanced exchange. Consideration is given to applications of the model to ocean outfall design and to the prediction of building ventilation flows. For natural ventilation, the theoretical model we derive for unbalanced exchange bridges the gap in the prediction of air flow rates between displacement flows, where the flow is unidirectional, and balanced exchange flows.

A versatile scanning method for volumetric measurements of velocity and density fields

Understanding turbulence in a stratified environment requires a detailed picture of both the velocity field and the density field. Experimentally, this represents a significant measurement challenge, especially when full three-dimensional data is needed to accurately characterise the turbulent fields. This paper presents a new approach to obtaining such data through well-resolved, near-instantaneous volume-spanning measurements. This is accomplished using a simultaneous particle image velocimetry and planar-laser-induced fluorescence scanning setup that is highly versatile while allowing precise overlapping light sheets in a volume. Moreover, the approach detailed here allows for complex scan sequences over varying volumes without the need for mechanical adjustment. To demonstrate the technique, preliminary results from the buoyancy-driven exchange flow through an inclined duct connecting two reservoirs of different density fluids are shown.

Buoyancy-driven exchange flow of immiscible yield-stress fluids in a vertical closed-ended container

We have numerically studied the buoyancy-driven flow of two immiscible yield-stress fluids in a vertical closed-ended pipe. Having assumed that the fluids of interest obey the bi-viscous model, we have relied on the finite-volume-method (FVM) in order to capture the late-time shape of the interface assumed to be initially flat. We have been able to recover a variety of well-established fluid-flow phenomena observed experimentally in this particular exchange flow for both Newtonian and yield-stress fluids. We have reached to the conclusion that, for immiscible fluids, the yield stress and/or the viscosity of the fluids can have a stabilizing effect on the interface under certain conditions. Our numerical results also suggest that surface tension and contact angle can play a key role in stabilizing the interface if properly chosen. A threshold curve is obtained for the “no-flow” zone and its variation with the fluids’ rheology and the contact angle is determined for bi-viscous fluids. In cases where fluids’ rheology and/or the wall's wetting property cannot be altered, shaking the vessel in the vertical direction is proposed as an active means for stabilizing the interface provided that its amplitude and frequency are within a certain range.

Experimental study on flow structure of buoyancy-driven exchange flow in horizontal pipes

The structure of buoyancy-driven exchange flow in a horizontal pipe was investigated experimentally by using brine water as the working fluid. The flow fields were visualized by laser-induced fluorescence (LIF), and the results were analyzed by a proper orthogonal decomposition (POD) to identify the variation of exchange-flow structure in the pipe. The LIF visualization enabled the investigation of the inflow and the purging conditions of the exchange flow in a horizontal pipe, which agreed closely without hysteresis, while they shifted to a higher Froude number regime in comparison with the vertical pipe conditions reported in the literature. The POD analysis of the exchange flow indicates that the fluctuating concentration level increased in the lower POD modes, owing to the periodic fluctuation of the exchange flow. The fluctuating concentration in the lowest POD mode increased at a lower Froude number, which reflects the periodic fluctuation of the interface on the lower wall and on both sides of the exchange flow. At higher Froude numbers, the exchange flow was suppressed and the fluctuating concentration level in the lower POD modes decreased. It was also found that the fluctuating velocity measured by particle image velocimetry (PIV) was highly magnified at the interface of the exchange flow of high Reynolds number, where the velocity gradient was highly generated owing to the interaction with the wall boundary.

Monodisperse particle-laden exchange flows in a vertical duct

We study buoyancy-driven exchange flow of two mixtures in a vertical narrow duct (two-dimensional channel as well as pipe) theoretically. While the light mixture is assumed always to be a pure fluid, the heavy mixture can be selected as either a pure or a particle-laden fluid. A small width-to-length ratio considered for the duct ($\unicode[STIX]{x1D6FF}\ll 1$) has been used as the perturbation parameter in developing a lubrication model (negligible inertia). In particular, we have adopted the methodology of Zhou et al. (Phys. Rev. Lett., vol. 94, 2005, 117803) for free-surface particle-laden film flows and extended it to a lock exchange system in confined geometry under the Boussinesq approximation. The resulting model is in the form of the classical Riemann problem and has been solved numerically using a robust total variation diminishing finite difference scheme. Both pure and particle-laden cases are investigated in detail. It is observed that the interface between the two fluids takes a self-similar shape at long times. In the case that both heavy and light fluids are pure, the dynamics of the flow is governed by two dimensionless quantities, namely the Reynolds number, $Re$, and the viscosity ratio, $\unicode[STIX]{x1D705}$, of the light and heavy fluids. The interpenetration of the heavy and light layers increases with $Re$ but decreases with $\unicode[STIX]{x1D705}$. Also, the heights of the heavy and light fronts change with $\unicode[STIX]{x1D705}$ but remain unchanged with $Re$. In the case of the particle-laden flow, however, four additional dimensionless parameters emerge, namely the initial volume fraction of particles, $\unicode[STIX]{x1D719}_{0}$, the ratio of particle diameter to duct width, $r_{p}$, and the density ratios of particles to carrying fluid, $\unicode[STIX]{x1D709}$, and of light fluid to carrying fluid, $\unicode[STIX]{x1D702}$. The effect of these parameters on the dynamics of the flow has been quantified through a systematic approach. In the presence of solid particles, the interface between the heavy and light layers becomes more curved compared to the case of pure fluids. This modification occurs due to the change of heavy mixture viscosity alongside the duct. Novel particle-rich zones are further discovered in the vicinity of the advancing heavy and light fronts. These zones are associated with different transport rates of the fluid and solid particles. The degree of particle enrichment remains the same with $Re$, is enhanced by $\unicode[STIX]{x1D705}$, $r_{p}$ and $\unicode[STIX]{x1D702}$, and is slightly diminished with $\unicode[STIX]{x1D719}_{0}$ and $\unicode[STIX]{x1D709}$. On the other hand, the stretched exchange zone between the heavy and light fronts grows with $r_{p}$, $\unicode[STIX]{x1D702}$ and $Re$, but decays with $\unicode[STIX]{x1D719}_{0}$, $\unicode[STIX]{x1D705}$ and $\unicode[STIX]{x1D709}$.

Non-isothermal buoyancy-driven exchange flows in inclined pipes

We study non-isothermal buoyancy-driven exchange flow of two miscible Newtonian fluids in an inclined pipe experimentally. The heavy cold fluid is released into the light hot one in an adiabatic small-aspect-ratio pipe under the Boussinesq limit (small Atwood number). At a fixed temperature, the two fluids involved have the same viscosity. Excellent qualitative and quantitative agreement is first found against rather recent studies in literature on isothermal flows where the driving force of the flow comes from salinity as opposed to temperature difference. The degree of flow instability and mixing enhances as the pipe is progressively inclined towards vertical. Similar to the isothermal limit, maximal rate of the fluids interpenetration in the non-isothermal case occurs at an intermediate angle, β. The interpenetration rate increases with the temperature difference. The degree of fluids mixing and diffusivity is found to increase in the non-isothermal case compared to the isothermal one. There has also been observed a novel asymmetric behavior in the flow, never reported before in the isothermal limit. The cold finger appears to advance faster than the hot one. Backed by meticulously designed supplementary experiments, this asymmetric behavior is hypothetically associated with the wall contact and the formation of a warm less-viscous film of the fluid lubricating the cold more-viscous finger along the pipe. On the other side of the pipe, a cool more-viscous film forms decelerating the hot less-viscous finger. Double diffusive effects associated with the diffusion of heat and mass (salinity) are further investigated. In this case and for the same range of inclination angles and density differences, the level of flow asymmetry is found to decrease. The asymmetric behaviour of the flow is quantified over the full range of experiments. Similar to the study of Salort et al. [“Turbulent velocity profiles in a tilted heat pipe,” Phys. Fluids 25(10), 105110-1–105110-16 (2013)] for tilted heat pipes, a small Richardson number of Ri≈0.05 is found, above which flow laminarization occurs. In terms of the dimensionless numbers of the problem, it is found that the interpenetrative speeds of the heavy and light fluid layers in non-isothermal and double-diffusive cases increase with the dimensionless temperature difference, rT, Atwood number, At, Grashof number, Gr, Reynolds number, Re, Nahme number, Na, and Péclet number, Pe but decreases with Prandtl number, Pr, and Brinkman number, Br.

Measurement of buoyancy-driven transient exchange flow rate across a thin horizontal ceiling vent of a non-adiabatic enclosure using a modified tracer-gas decay method

ABSTRACTHorizontal openings are important pathways for mass and energy transfer between different zones in buildings or between a building and the outdoor environment. The temperature difference between indoor air and envelopes and its resulting heat transfer is common in buildings, especially in ones with transient ventilation. The transient, buoyancy-driven ventilation flow rates across a thin horizontal ceiling vent of a non-adiabatic enclosure are experimentally studied. Rather than using brine/water to imitate the buoyancy difference between indoor and outdoor fluids as been done in most previous studies, this study employs air as the working fluid and thus the effects of heat transfer between indoor air and envelopes can be taken into account. To measure the buoyancy-driven exchange flow rate that is dependant of the difference between indoor and outdoor temperature, a modified tracer-gas concentration decay method is proposed. In this method, the ventilation rate is determined by the combination of decay curve of tracer-gas concentration and that of temperature-rise. It is shown that the heat transfers at the enclosure envelopes have an influence on the exchange flow rate of a ceiling vent. Both a case with big rate of heat transfer from envelopes to indoor air and the one from indoor air to envelopes could produce dimensionless flow rates as twice as the model predictions that ignore the heat transfers at the envelopes (e.g. the Epstein's formula derived from salt-water experiments). It is also shown that is independent of the dimensionless temperature rise of indoor air .

Buoyancy-driven exchange flow between two adjacent building zones connected with top and bottom vents

We examine the buoyancy driven flow between two adjacent building zones separated by top and bottom vents with and without a source of buoyancy. In the absence of a source, by opening the vents, the pressure difference between the two zones drives the exchange flow. When there is an isolated source of buoyancy, two distinct flow regimes are defined: a ventilation-dominated regime in which the flow that is generated by the initial temperature difference, and a source-dominated regime in which the flow is principally governed by the source. Here, we restrict attention to the former case. Thus in each zone our analytical model assumes a horizontal interface that separates the upper layer, which is comprised of fluid having the initial density of the light zone, from the lower layer, comprised of fluid whose density is larger than that of the light zone. The transient evolution of such interfaces along with the stratification and buoyancy in each zone are predicted. Attention is focused on the influence of the effective area, source buoyancy flux and the time during which the source is switched on. Because the interface elevations change significantly at early times, a non-monotonic behavior is predicted for the evolution of stratification in the light zone. We also find that the terminal stratification and buoyancy in both zones are not impacted by the number of connecting vents; however, this number has a strong influence on how quickly steady state is achieved. Similitude laboratory experiments help to corroborate key predictions.

CFD benchmark on hydrogen release and dispersion in a ventilated enclosure: Passive ventilation and the role of an external wind

In the framework of the H2FC project a Computational Fluid Dynamics (CFD) benchmark was performed to study the release and dispersion of hydrogen in a naturally ventilated enclosure with one vent. The benchmark involved comparing CFD model predictions with measurements from an experiment carried out by the Health and Safety Laboratory (HSL). A total of 28 experiments were performed by HSL using their 31 m3 enclosure and test 25 was chosen for this benchmark. In test 25, hydrogen was released vertically upwards through a 0.55 mm diameter nozzle located 0.5 m above the center of the floor of the enclosure. The release was sonic with a volume release rate of 169 NL/min. HSL's experimental facility is exposed to naturally varying wind conditions. During test 25 the wind blew from the opposite side of the enclosure to the vent at an average speed of 2.6 m/s. Three H2FC partners participated in the benchmark, with three different CFD codes, (ANSYS CFX, ADREA-HF and ANSYS FLUENT) and two different turbulence models (standard k-ε and LES with Smagorinski-Lilly model). In general, satisfactory agreement was found between predicted and measured hydrogen concentrations. However, the k-ε model overestimates the hydrogen concentration at most of the sensors, while the LES model underestimates it, especially at the bottom sensors. Two additional cases with lower wind speed and with no wind were examined. They show that the presence of an external wind can influence the hydrogen dispersion inside an enclosure with one vent. In particular, they show that an external wind can disrupt the buoyancy-driven exchange flow through a vent and lead to less efficient ventilation.

CFD ANALYSIS OF FLOW THROUGH HORIZONTAL CEILING OPENING DUE TO BUOYANCY AND PRESSURE DIFFERENCE

The buoyancy driven exchange flow through the large openings in horizontal partitions occurs in many practical situations such as in enclosed regions with a ceiling opening and a heat source such as fire. The density difference between two compartments arises partly due to difference in composition and partly from the difference in temperature. A heavier fluid located on the top of a lighter fluid and separated by a horizontal vent constitutes a gravitationally unstable system. Horizontal vents produce flow, which are unstable with irregular oscillatory behavior. However, when lower compartment is slightly pressurized the flow becomes stable and unidirectional. A numerical study has been performed to characterize the bi-directional flow and transition to unidirectional flow through a horizontal vent in an enclosure, due to differences in pressure and density across the vent. Fresh and salt water has been considered as working fluids to create density difference across the vent with a pressure field imposed in the lower compartment. The pressure in the lower region was increased to find the critical pressure corresponding to transition to unidirectional from bi-directional flow. Unsteady, 2D axisymmetric, incompressible Navier–Stokes equation along with species, turbulence and continuity equation have been solved with finite volume method using the in-house computational fluid dynamics (CFD) code. Several cases were examined to compute the critical pressure for various density differences for low opening aspect ratio. The code has been validated with reported experiments and used to simulate various other practical cases occurred during fire induced flow through such openings.

On the flow of buoyant fluid injected into an aquifer with a background flow

AbstractWe study the dispersal of a plume of incompressible buoyant fluid injected into a confined inclined aquifer in which there is a background flow. We assume that, to prevent pressure buildup in the system, there is an outflow from the aquifer, with flux equal to the injection flux, through a producing well. Using the method of characteristics, we identify that the trajectory of the plume of injected fluid depends on the magnitudes of both the injection flux ${Q}_{I} $ and the background aquifer flux ${Q}_{A} $ relative to the buoyancy-driven exchange flow of injected and original fluid within the aquifer ${Q}_{E} $, on the direction of the background aquifer flow, and on whether the producing well lies upslope or downslope from the injecting well. We find the values of the controlling parameters ${Q}_{I} / {Q}_{E} $ and ${Q}_{A} / {Q}_{E} $ for which all injected fluid flows up-dip, for which the injected fluid partitions into a component moving up-dip and a component moving down-dip, and for which all injected fluid flows down-dip. A key learning from the analysis is that there may be very different plume trajectories when a buoyant fluid is injected into a confined, inclined aquifer, and prediction of the trajectory depends on knowledge of the background flow as well as the injection rate and location of the producing wells. In the process of ${\mathrm{CO} }_{2} $ sequestration, this range of initial plume geometries can inform analysis of longer-term geological storage and assessment of the risk of activating different possible leakage pathways to the surface.

Circulation of bubbly magma and gas segregation within tunnels of the potential Yucca Mountain repository

Following an intersection of rising magma with drifts of the potential Yucca Mountain nuclear waste repository, a pathway is likely to be established to the surface with magma flowing for days to weeks and affecting the performance of engineered structures located along or near the flow path. In particular, convective circulation could occur within magma-filled drifts due to the exsolution and segregation of magmatic gas. We investigate gas segregation in a magma-filled drift intersected by a vertical dyke by means of analogue experiments, focusing on the conditions of sustained magma flow. Degassing is simulated by electrolysis, producing micrometric bubbles in viscous mixtures of water and golden syrup, or by aerating golden syrup, producing polydisperse bubbly mixtures with 40% of gas by volume. The presence of exsolved bubbles induces a buoyancy-driven exchange flow between the dyke and the drift that leads to gas segregation. Bubbles segregate from the magma by rising and accumulating as a foam at the top of the drift, coupled with the accumulation of denser degassed magma at the base of the drift. Steady-state influx of bubbly magma from the dyke into the drift is balanced by outward flux of lighter foam and denser degassed magma. The length and time scales of this gas segregation are controlled by the rise of bubbles in the horizontal drift. Steady-state gas segregation would be accomplished within hours to hundreds of years depending on the viscosity of the degassed magma and the average size of exsolved gas bubbles, and the resulting foam would only be a few cm thick. The exchange flux of bubbly magma between the dyke and the drift that is induced by gas segregation ranges from 1 m3 s−1, for the less viscous magmas, to 10−8 m3 s−1, for the most viscous degassed magmas, with associated velocities ranging from 10−1 to 10−9 m s−1 for the same viscosity range. This model of gas segregation also predicts that the relative proportion of erupted degassed magma, that could potentially carry and entrain nuclear waste material towards the surface, would depend on the value of the dyke magma supply rate relative to the value of the gas segregation flux, with violent eruption of gassy as well as degassed magmas at relatively high magma supply rates, and eruption of mainly degassed magma by milder episodic Strombolian explosions at relatively lower supply rates.

Gas segregation in dykes and sills

Many basaltic volcanoes emit a substantial amount of gas over long periods of time while erupting relatively little degassed lava, implying that gas segregation must have occurred in the magmatic system. The geometry and degree of connectivity of the plumbing system of a volcano control the movement of magma in that system and could therefore provide an important control on gas segregation in basaltic magmas. We investigate gas segregation by means of analogue experiments and analytical modelling in a simple geometry consisting of a vertical conduit connected to a horizontal intrusion. In the experiments, degassing is simulated by electrolysis, producing micrometric bubbles in viscous mixtures of water and golden syrup. The presence of exsolved bubbles induces a buoyancy-driven exchange flow between the conduit and the intrusion that leads to gas segregation. Bubbles segregate from the fluid by rising and accumulating as foam at the top of the intrusion, coupled with the accumulation of denser degassed fluid at the base of the intrusion. Steady-state influx of bubbly fluid from the conduit into the intrusion is balanced by outward flux of lighter foam and denser degassed fluid. The length and time scales of this gas segregation are controlled by the rise of bubbles in the horizontal intrusion. Comparison of the gas segregation time scale with that of the cooling and solidification of the intrusion suggests that gas segregation is more efficient in sills (intrusions in a horizontal plane with typical width:length aspect ratio 1:100) than in horizontally-propagating dykes (intrusions in a vertical plane with typical aspect ratio 1:1000), and that this process could be efficient in intermediate as well as basaltic magmas. Our investigation shows that non-vertical elements of the plumbing systems act as strong gas segregators. Gas segregation has also implications for the generation of gas-rich and gas-poor magmas at persistently active basaltic volcanoes. For low magma supply rates, very efficient gas segregation is expected, which induces episodic degassing activity that erupts relatively gas-poor magmas. For higher magma supply rates, gas segregation is expected to be less effective, which leads to stronger explosions that erupt gas-rich as well as gas-poor magmas. These general physical principles can be applied to Stromboli volcano and are shown to be consistent with independent field data. Gas segregation at Stromboli is thought likely to occur in a shallow reservoir of sill-like geometry at 3.5 km depth with exsolved gas bubbles 0.1–1 mm in diameter. Transition between eruptions of gas-poor, high crystallinity magmas and violent explosions that erupt gas-rich, low crystallinity magmas are calculated to occur at a critical magma supply rate of 0.1–1 m 3 s − 1 .

Numerical Simulations on Dust-Air Two-Phase Flows in Fusion Experimental Reactors During Loss-of-Vacuum-Accident Events

Dust-air two-phase flow characteristics in a fusion experimental reactor during a loss-of-vacuum-accident (LOVA) event were analyzed numerically by three dimensional simulations using a newly developed thermal-hydraulic analysis code. Physical models on the motion of dust were considered to resolve the dust mobilization conveying by the fluid. Air ingress behavior through a breach at the LOVA event was calculated by using compressible Navier-Stokes equations. It was predicted quantitatively from the results of the present numerical study that the dust mobilization receives strongly the effect of the breach size and the fraction of the mobilized dust is determined by a circulating flow and buoyancy-driven exchange flow which are generated in a vacuum vessel of the fusion experimental reactor after the LOVA event.

Density stratification from buoyancy-driven exchange flow through horizontal partitions in a liquid tank

This paper presents an experimental study of the transient density stratification from buoyancy-driven and forced-convection flows through horizontal partitions in a liquid tank. When strongly stratified, an enclosure's ambient density distributions can be considered as one-dimensional, with negligible horizontal gradients except in narrow regions around buoyant jets. The vertical density distribution was measured by sheet laser light passing through the corner of the tank, using the change of light reflection index associated with the local fluid component concentration and density. The results of these experiments gives an important information to improve the modeling of transient stratification evolution induced by jets and plumes in large enclosures, as well as the prediction of exchange flow rates through horizontal openings between inter-connected compartments.

4. Reyleigh-taylor instability of buoyancy driven exchange flow

4. Reyleigh-taylor instability of buoyancy driven exchange flow